Semantic Interpolation in Implicit Models

By Yannic Kilcher, Aurélien Lucchi, Thomas Hofmann. 2017

18 citations recorded by the source corpus.

In implicit models, one often interpolates between sampled points in latent space. As we show in this paper, care needs to be taken to match-up the distributional assumptions on code vectors with the geometry of the interpolating paths. Otherwise, typical assumptions about the quality and semantics of in-between points may not be justified. Based on our analysis we propose to modify the prior code distribution to put significantly more probability mass closer to the origin. As a result, linear interpolation paths are not only shortest paths, but they are also guaranteed to pass through high-density regions, irrespective of the dimensionality of the latent space. Experiments on standard benchmark image datasets demonstrate clear visual improvements in the quality of the generated samples and exhibit more meaningful interpolation paths.

Subfields: Computer Vision and Pattern Recognition, Generative Models

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Generative Models

Citations in the corpus

ImageBART: Bidirectional Context with Multinomial Diffusion for Autoregressive Image Synthesis (2021, background): “This makes them easy to sample, interpolate [41, 34] and modify [74].”
GAN Steerability without optimization (2020, methodology): “To avoid this, some methods proposed to replace the popular Gaussian latent space distribution by other priors (Kilcher et al., 2018), or to optimize the generator together with the latent space (Bojanowski et al., 2018).”
Distribution-Interpolation Trade off in Generative Models (2018, methodology): “Nonetheless, reducing the consequences of the distribution mismatch is still a popular research topic (Agustsson et al., 2017; Kilcher et al., 2017; Arvanitidis et al., 2017).”